During two half-days, March 12 and April 6, LINK-SIC arranged workshops with presentation of the theoretical background of Reinforcement learning and hands-on activities.

The presentations were given by Farnaz Adib Yaghmaie and Lennart Ljung, division of Automatic control at the department of Electrical engineering.

Handout for the workshop: A Crash Course on Reinforcement Learning (External page)

Notes about Coding for Reinforcement Learning Workshop (PDF)

Code for the Workshop (External page)

Schedule LINK-SIC Core Competence Workshops on Reinforcement Learning

Schedule for Day 1, 2021-03-12 

13:15-13:25 Introduction to Reinforcement learning (PDF)
13:25-13:30 Discussion
13:30-13:40 Markov Decision Process (PDF)
13:40-13:50 Break 
13:50-14:10 Policy Gradient (PDF)
14:10-14:20 Break
14:20-14:30 Policy Gradient on Cartpole (PDF)
14:30-14:50 Coding and Discussion
14:50-15:05 Break 
15:05-15:15 Policy Gradient on Linear Quadratic (PDF)
15:15-15:35 Coding and Discussion
15:35-15:50 Break 
15:50-16:20 Model Building (PDF)

Schedule for Day 2, 2021-04-06

13:15-13:25 Recap of Reinforcement learning and MDP (PDF)
13:25-13:30 Discussion for those not participated in Day 1
13:30-13:50 Q Learning (PDF)
13:50-14:00 Break 
14:00-14:10 Presentation Q learning on Cartpole (PDF)
14:10-14:30 Coding and Discussion
14:30-14:45 Break 
14:45-14:55 Q learning on Linear Quadratic (PDF)
14:55-15:15 Coding and Discussion
15:15-15:30 Break 
15:30-16:15 Panel Discussion (PDF)